Table 6 Results of the test of potential moderators.

From: Revisiting the Porter hypothesis: a multi-country meta-analysis of the relationship between environmental regulation and green innovation

Group

Effect size and 95% confidence interval

Heterogeneity

Number of studies

Point estimate

Sample size

95% CI

Q-value

p-value

Overall regulation and number of patents

Country type

Developing country

12

0.10

4144

[−0.18, 0.37]

0.98

0.32

Developed country

3

0.37

706,990

[−0.10, 0.71]

Level of analysis

Firm

6

0.09

707,931

[−0.00, 0.19]

13.96

0.003**

City

3

−0.36

3035

[−0.83, 0.41]

Province

3

0.29

90

[−0.13, 0.62]

Country

3

0.68

78

[0.40, 0.85]

Overall regulation and pollution-based indices

Level of analysis

Firm

1

−0.80

21

[−0.92, −0.56]

30.02

0.000***

Province

1

0.51

30

[0.19, 0.74]

Command and control regulation and number of patents

Country type

Developing country

17

0.14

20,369

[0.06, 0.22]

4.08

0.04*

Developed country

4

0.39

38,447

[0.16, 0.58]

Level of analysis

Firm

14

0.10

58,013

[0.04, 0.16]

10.18

0.02*

City

1

0.20

266

[0.08, 0.31]

Province

3

0.25

450

[0.16, 0.33]

Country

3

0.62

87

[0.01, 0.89]

Market incentive regulation and number of patents

Level of analysis

Firm

1

0.13

573

[0.04, 0.21]

36.49

0.000***

City

1

−0.07

3557

[−0.11, −0.04]

Province

2

0.17

420

[0.07, 0.26]

  1. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001 (significance levels based on two-tailed tests).